How does the TikTok algorithm work?

TikTok reveals details of how TikTok algorithm works

How does the TikTok algorithm work? The company gave more details on the algorithmic insights of the app, to dispel myths and rumors about the company.

The virality of the information disseminated on the application has found its counterpart in the news in recent weeks. It’s an understatement to say that TikTok is at the center of the conversation, between its ban in India and Trump’s renewed belching. Threatened with ban also in the United States, discussions are going well on a potential resale of the application. We recently mentioned a new regulation that came into force in China aimed at restricting companies located in the country in their exports and their technological transactions, calling into question, in fact, the relevance of acquisition of TikTok without its algorithms.

It is on these that the company has just given new elements to a handful of carefully selected journalists. So how does TikTok’s algorithm work?

 

How does the TikTok algorithm work?

 

How does the TikTok algorithm work?

 

While TikTok struggles to stop the broadcasting of a video or a person commits suicide, doped by his algorithm, the failings of the latter are pointed out. New insights into how the app’s algorithm works were reported by Axios in an article published on September 10.

IN THE SAME CATEGORY

TikTok executives explain that when logging into the application for the first time, the user is prompted with 8 very different already popular videos, presenting different trends, music, and themes. Subsequently, the algorithm will continue to push videos based on the content with which the user engages and their use of the platform. As we reported to you in an article on the secrets of the TikTok algorithm, engagement is built around actions like:

– interactions with the video (comments, likes, subscriptions, and shares;

– the duration of the video consultation ( completion rates ).

 

But the data collected is broader with information on the video viewed itself: the presence of captions, sounds, and music, hashtags. The user’s language, country, or device type are also taken into account.

And that’s where it gets interesting, because the self-learning algorithm with the enrichment of user data, provides more and more relevant and personalized content.

 

From engagement to user cluster mapping

 

Once the necessary user data is collected, TikTok will be able to identify the preferences of a person and create clusters of similar users. At the same time, the application’s algorithm groups the videos by similar themes. Machine learning does the rest: based on pre-established tastes through user engagement and segmentation into similar user groups, TikTok will be able to deliver the right videos to the right profile.

Unlike other social networks based on the popularity of accounts and the subscription to other users, TikTok takes the party to deliver the content on criteria of relevance and virality. This is how user videos with a small community of followers can reach a titanic number of views.

 

An algorithm that also produces “filter bubbles”

Does the term sound familiar to you? Filter bubbles or filter bubbles have been mentioned for “old” social networks like Facebook and Twitter with the excesses that we know: conspiracy theories, fake news… It is paradoxical for an algorithm that works differently on TikTok, but also on the application, users can find themselves confirmed and reinforced in their prior preferences.

TikTok concedes that the dark side of sensing user preferences is that using user preferences leads to tapping on those same preferences by tuperwarizing their ability to broaden the spectrum of content they might discover and like. In this sense, if the company claims to study filter bubbles, for the time being, only its desire to continually offer fresh, non-redundant, and viral content to the user seems capable of opening it up to other sensibilities. in terms of content.

How does the TikTok algorithm work? The company gave more details on the algorithmic insights of the app, to dispel myths and rumors about the company.

 

The virality of the information disseminated on the application has found its counterpart in the news in recent weeks. It’s an understatement to say that TikTok is at the center of the conversation, between its ban in India and Trump’s renewed belching. Threatened with ban also in the United States, discussions are going well on a potential resale of the application. We recently mentioned a new regulation that came into force in China aimed at restricting companies located in the country in their exports and their technological transactions, calling into question, in fact, the relevance of acquisition of TikTok without its algorithms.

It is on these that the company has just given new elements to a handful of carefully selected journalists. So how does TikTok’s algorithm work?

 

How does the TikTok algorithm work?

 

While TikTok struggles to stop the broadcasting of a video or a person commits suicide, doped by his algorithm, the failings of the latter are pointed out. New insights into how the app’s algorithm works were reported by Axios in an article published on September 10.

IN THE SAME CATEGORY

 

TikTok executives explain that when logging into the application for the first time, the user is prompted with 8 very different already popular videos, presenting different trends, music, and themes. Subsequently, the algorithm will continue to push videos based on the content with which the user engages and their use of the platform. As we reported to you in an article on the secrets of the TikTok algorithm, engagement is built around actions like:

– interactions with the video (comments, likes, subscriptions, and shares;

– the duration of the video consultation ( completion rates ).

 

But the data collected is broader with information on the video viewed itself: the presence of captions, sounds, and music, hashtags. The user’s language, country, or device type are also taken into account.

And that’s where it gets interesting, because the self-learning algorithm with the enrichment of user data, provides more and more relevant and personalized content.

Also, Read How Much Do TikTokers Earn..?

From engagement to user cluster mapping

Once the necessary user data is collected, TikTok will be able to identify the preferences of a person and create clusters of similar users. At the same time, the application’s algorithm groups the videos by similar themes. Machine learning does the rest: based on pre-established tastes through user engagement and segmentation into similar user groups, TikTok will be able to deliver the right videos to the right profile.

Unlike other social networks based on the popularity of accounts and the subscription to other users, TikTok takes the party to deliver the content on criteria of relevance and virality. This is how user videos with a small community of followers can reach a titanic number of views.

 

An algorithm that also produces “filter bubbles”

Does the term sound familiar to you? Filter bubbles or filter bubbles have been mentioned for “old” social networks like Facebook and Twitter with the excesses that we know: conspiracy theories, fake news… It is paradoxical for an algorithm that works differently on TikTok, but also on the application, users can find themselves confirmed and reinforced in their prior preferences.

TikTok concedes that the dark side of sensing user preferences is that using user preferences leads to tapping on those same preferences by tuperwarizing their ability to broaden the spectrum of content they might discover and like. In this sense, if the company claims to study filter bubbles, for the time being, only its desire to continually offer fresh, non-redundant, and viral content to the user seems capable of opening it up to other sensibilities. in terms of content.

 

1 thought on “How does the TikTok algorithm work?”

Leave a Comment

Your email address will not be published. Required fields are marked *

Show Buttons
Hide Buttons